[go: up one dir, main page]

CN1956555B - Apparatus and method for processing 3d picture - Google Patents

Apparatus and method for processing 3d picture Download PDF

Info

Publication number
CN1956555B
CN1956555B CN2006101639357A CN200610163935A CN1956555B CN 1956555 B CN1956555 B CN 1956555B CN 2006101639357 A CN2006101639357 A CN 2006101639357A CN 200610163935 A CN200610163935 A CN 200610163935A CN 1956555 B CN1956555 B CN 1956555B
Authority
CN
China
Prior art keywords
image
eye image
eye
parallax
disparity
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Fee Related
Application number
CN2006101639357A
Other languages
Chinese (zh)
Other versions
CN1956555A (en
Inventor
河泰铉
具宰必
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Samsung Electronics Co Ltd
Original Assignee
Samsung Electronics Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Samsung Electronics Co Ltd filed Critical Samsung Electronics Co Ltd
Publication of CN1956555A publication Critical patent/CN1956555A/en
Application granted granted Critical
Publication of CN1956555B publication Critical patent/CN1956555B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00Stereoscopic video systems; Multi-view video systems; Details thereof
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106Processing image signals
    • H04N13/144Processing image signals for flicker reduction
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106Processing image signals
    • H04N13/122Improving the 3D impression of stereoscopic images by modifying image signal contents, e.g. by filtering or adding monoscopic depth cues
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N13/00Stereoscopic video systems; Multi-view video systems; Details thereof
    • H04N13/10Processing, recording or transmission of stereoscopic or multi-view image signals
    • H04N13/106Processing image signals
    • H04N13/128Adjusting depth or disparity
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N2213/00Details of stereoscopic systems
    • H04N2213/002Eyestrain reduction by processing stereoscopic signals or controlling stereoscopic devices

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)

Abstract

提供了一种图像处理设备和方法,其可以调整输入的三维(3D)图像的视差,并且稳定地显示3D图像。图像处理设备包括:接收3D图像并且将3D图像划分为左眼图像和右眼图像的图像划分单元;估计左眼图像和右眼图像之间的视差、测量估计的视差发生的频率、并且产生视差和频率的频率分布图的频率分布图产生单元;使用产生的频率分布图确定左眼图像和右眼图像的水平运动值的水平运动值确定单元;和基于确定的水平运动值来水平地移动左眼图像和右眼图像的水平运动单元。图像处理设备可以控制图像的收敛以加强3D效果和减少眼睛疲劳。

Figure 200610163935

Provided are an image processing apparatus and method that can adjust parallax of an input three-dimensional (3D) image and stably display the 3D image. The image processing apparatus includes: an image division unit that receives a 3D image and divides the 3D image into a left-eye image and a right-eye image; estimates parallax between the left-eye image and the right-eye image, measures frequency of occurrence of the estimated parallax, and generates the parallax and a frequency histogram generating unit of frequency histograms; a horizontal motion value determination unit that determines horizontal motion values of the left-eye image and the right-eye image using the generated frequency histogram; and moves the left side horizontally based on the determined horizontal motion values Horizontal motion units for eye and right eye images. The image processing device can control the convergence of the image to enhance the 3D effect and reduce eye fatigue.

Figure 200610163935

Description

3-D view treatment facility and method
Technical field
The present invention relates to be used for handle the equipment and the method for three-dimensional (3D) image, relate in particular to the parallax (disparity) of the 3D rendering that is used to adjust input and stably show the image processing equipment and the method for 3D rendering.
Background technology
Traditional three-dimensional (3D) display shows 3-D view by the image projection using two dimension (2D) flat-panel screens and instrument (such as the little polarizer of membrane type (film-type micropolarizer), parallax barrier (parallax barrier) and biconvex lens (lenticular lens)) and will have different parallaxes to user's left eye and right eye.This 3D display of number according to the view that shows on it can be divided into solid (stereoscopic) 3D display and many views (multiview) 3D display.Stereoscopic 3 d display shows the image with two different views, and many views 3D display shows to have two images with top view.
In order to produce 3D rendering, real image is by the stereoscopic camera that comprises two incident lens or comprise many view cameras photograph (filmed) of some incident lens.Alternately, solid or the multi-view image content figure that can use a computer produces.
When using the 3D camera to produce 3D rendering maybe when the figure that uses a computer produces picture material, need to consider people's visual signature.Yet, when display environment is different from the generation environment, for example when the 3D rendering that uses many view camera to produce shows on stereoscopic 3 d display or when the displays of different sizes were used for generation and demonstration 3D rendering, traditional 3D display can not show 3D rendering as required.
Summary of the invention
One aspect of the present invention provides image processing equipment and method, and it can be used for strengthening three-dimensional (3D) effect and reduce eye fatigue being used for 3D stereo display to use the image that is obtained by many view camera.
One aspect of the present invention provides image processing equipment and method, and it can be used for the convergence (convergence) that comes the control chart picture by the parallax of adjusting image at the receiving terminal that receives and show 3D rendering.
According to one aspect of the invention, provide image processing equipment to comprise: to receive three-dimensional (3D) image and 3D rendering is divided into the image division unit of left-eye image and eye image; The histogram generation unit of the histogram (histogram) of the parallax between the corresponding units image of estimation left-eye image and eye image, the frequency of measuring the estimating disparity generation and generation parallax and frequency; Use the histogram that produces to determine the horizontal movement value determining unit of the horizontal movement value of left-eye image and eye image; Based on flatly mobile left-eye image of determining of horizontal movement value and at least one the horizontal movement unit in the eye image.
The threshold value that horizontal movement value determining unit calculated threshold and use are calculated is determined the horizontal movement value, and described threshold value makes that the estimated rate of parallax is a positive parallax in the histogram.Described estimated rate is the steady state value that is confirmed as making that at least one backward three-dimensional viewing in mobile left-eye image and the eye image has optimal depth consciousness.
Horizontal movement value determining unit can determine that corresponding to half value of threshold value be the horizontal movement value.
Thereby the horizontal movement unit can be based on the mobile left-eye image in threshold level ground and eye image and adjustment parallax.
Described equipment can further comprise user interface section, it receives estimated rate or threshold value and sends estimated rate or threshold value to horizontal movement value determining unit from the user by user input signal, and wherein horizontal movement value determining unit uses the estimated rate or the threshold value that are received to determine the horizontal movement value.Described equipment can further comprise the 3D stereoscopic display unit, and it shows from the left-eye image and the eye image of the adjusted parallax of horizontal movement unit output.
According to another aspect of the present invention, provide image processing method to comprise: to receive three-dimensional (3D) image and 3D rendering is divided into left-eye image and eye image; Parallax between the corresponding units image of estimation left-eye image and eye image is measured the frequency that estimating disparity takes place, and produces the histogram of parallax and frequency; Use the histogram that produces to determine the horizontal movement value of left-eye image and eye image; Based on the flatly mobile left-eye image of determining of horizontal movement value and in the eye image at least one to adjust parallax.The determining of described horizontal movement value comprises: calculated threshold, this threshold value make that the estimated rate of parallax is a positive parallax in the histogram; And use the threshold value of being calculated to determine the horizontal movement value.Described estimated rate is the steady state value that is confirmed as making that at least one backward three-dimensional viewing in mobile left-eye image and the eye image has optimal depth consciousness.
According to another aspect of the present invention, provide a kind of image processing equipment to comprise: image division unit, it receives image and is first image and second image with image division; The histogram generation unit, it estimates the parallax between the corresponding units image of first image and second image, measures the frequency that the parallax estimated takes place, and produces the histogram of parallax and frequency; The motion value determining unit, it uses the histogram that produces to determine the motion value of first image and second image; And moving cell, it moves in first image and second image at least one based on the motion value of determining, and, so adjust the parallax of image.Wherein, described motion value determine comprise: calculated threshold, this threshold value make that the estimated rate of parallax is a positive parallax in the histogram; And use the threshold value of being calculated to determine described motion value.Described estimated rate is to be confirmed as the steady state value that feasible at least one backward three-dimensional viewing that moves in first image and second image has optimal depth consciousness.
According to another aspect of the present invention, provide a kind of image processing method, comprising: receive image, and be first image and second image image division; Estimate the parallax between the corresponding units image of first image and second image, measure the frequency that the parallax estimated takes place, and produce the histogram of parallax and frequency; Use the histogram that produces to determine the motion value of first image and second image; And move in first image and second image at least one based on the motion value of determining, and, the parallax of image therefore adjusted.Wherein, described motion value determine comprise: calculated threshold, this threshold value make that the estimated rate of parallax is a positive parallax in the histogram; And use the threshold value of being calculated to determine described motion value.Described estimated rate is to be confirmed as the steady state value that feasible at least one backward three-dimensional viewing that moves in first image and second image has optimal depth consciousness.
According to another aspect of the present invention, provide a kind of computer-readable medium storing that has write down the program that is used for the carries out image processing method on it.
Description of drawings
Describe exemplary embodiment wherein in conjunction with the drawings in detail, above-mentioned and other feature and advantage of the present invention will become more apparent, in the accompanying drawings:
Figure 1A-1C illustrates the type of stereoscopic camera;
Fig. 2 illustrates the problem that produces when using the toed-in stereoscopic camera to carry out shooting operation;
Fig. 3 illustrates the structure of many view camera;
Fig. 4 illustrates the application of multi-view video coding;
Fig. 5 A-5C illustrates according to an embodiment of the invention left-eye image and the eye image by parallel many view cameras photograph;
Fig. 6 A-6C illustrates according to an exemplary embodiment of the present invention and produces the stable and stereoscopic treatment of picture with being moved to the left to the right respectively by left-eye image and eye image with accompanying drawing 5;
Fig. 7 illustrates according to an exemplary embodiment of the present invention the disparity estimation (DE) based on piece (block-based);
Fig. 8 illustrates the histogram of parallax between the left-eye image and eye image according to an exemplary embodiment of the present invention;
Fig. 9 A-9C illustrates three kinds of histograms of parallax between the left-eye image in the multi-view image that obtains by parallel (parallel) many view camera and eye image according to an exemplary embodiment of the present invention;
Figure 10 A-10C illustrates according to an exemplary embodiment of the present invention based on the left-eye image of horizontal movement value and moving horizontally of eye image, and described horizontal movement value uses the parallax histogram to determine;
Figure 11 is the block diagram of image processing equipment according to an embodiment of the invention; With
Figure 12 illustrates the flow chart of image processing method according to an embodiment of the invention.
Embodiment
Now will more comprehensively describe the present invention in conjunction with the accompanying drawings, wherein show exemplary embodiment of the present.Yet the present invention can be presented as a lot of different forms and should not be interpreted as is to be limited to the embodiment that wherein sets forth; More precisely, these embodiment are provided in case this openly will be completely and completely, and send thought of the present invention to those skilled in the art fully.
The structure of the camera that produces real image (real image) is described in conjunction with Fig. 1 and 3 now.
Figure 1A-1C illustrates the type of stereoscopic camera.Figure 1A illustrates parallel stereoscopic camera.The aligning of parallel stereoscopic camera is simple.Yet, because parallel stereoscopic camera do not have convergence function, consider the people visual signature it can not photographic images.Therefore, captured image can not be as three-dimensional (3D) image.Figure 1B illustrates the toed-in stereoscopic camera.The Toed-in stereoscopic camera has convergence function, and has critically reflected people's visual signature.Yet the toed-in stereoscopic camera has the problem shown in the accompanying drawing 2.
Accompanying drawing 2 illustrates the problem that produces when using the toed-in stereoscopic camera shown in the accompanying drawing 1B to carry out shooting operation.
When rectangular object 21 was positioned between the camera axis of the left and right sides camera shown in Fig. 2, the image of taking by a left side and right camera 23 and 25 was out of shape as shown in Figure 2 respectively.When image 23 and 25 was used for 3D rendering, image 23 and 25 this distortion caused eye fatigue.
Fig. 1 C illustrates mixing (hybrid) stereoscopic camera of aiming in the mode that prevents parallel stereoscopic camera that Figure 1A and 1B illustrate respectively and toed-in stereoscopic camera shortcoming.In mixing stereoscopic camera, lens unit and charge-coupled device (CCD) unit can move individually.Therefore, can control convergence by the position of adjusting CCD unit and lens.
Fig. 3 illustrates the structure of many view camera.
The conventional method of taking multi-view image is used many view camera of structure as shown in Figure 3.Carry out in the standardization at the 3D of Motion Picture Experts Group (MPEG) audio frequency and video (3DAV), the image of the many view cameras photograph by accompanying drawing 3 is as standard picture.Stereoscopic camera can have the structure shown in the accompanying drawing 1C.Yet under the situation that multi-view video is taken, the convergence of controlling each image is actually infeasible.
Fig. 4 illustrates the application of multi-view video coding.
With reference to figure 4, the image of exporting from many view camera of being made up of N camera is compressed and stores or transmit multi-view coded processing, rebuilds in many views decodings are handled then.Reconstructed image can be according to the demonstration that should be used for of display device.That is, the image sequence of view is presented on the 2D display.For 3D stereo display, two views are selected and be used for showing 3D rendering on the 3D three-dimensional display.Show that for the many views of 3D at least two views are selected and be used for showing on the 3D multi-view display 3D multi-view image.When two image sequences were used for showing 3D rendering on the 3D three-dimensional display, the 3D three-dimensional display had the same problem of parallel stereoscopic camera shown in Figure 1A.Therefore, the image that obtains by many view camera must correctly be handled to show on the 3D three-dimensional display.
Fig. 5 A-5C illustrates according to an embodiment of the invention left-eye image and the eye image by parallel many view cameras photograph.When supposition was chosen two views and is used for 3D stereo display from the image sequence by parallel many view camera outputs, the phenomenon shown in Fig. 5 A-5C will appear.
Fig. 5 A illustrates left-eye image, and Fig. 5 B illustrates eye image.Fig. 5 C is illustrated in the left-eye image of Fig. 5 A on the eye image that is superimposed upon Fig. 5 B in the plane to show the parallax between them.Suppose that just there is positive parallax in target (objects) when eye image when being present in the right of same target of left-eye image.In this case, the image of all targets has positive parallax in the superimposed images shown in Fig. 5 C.In other words, we can say because the feature of parallel camera does not exist convergence point or convergence point to be present in outside the screen.
When the superimposed images shown in Fig. 5 C showed on the 3D three-dimensional display, the image of all targets seemed outstanding from screen in the overlapping image.This parallax, just, the degree of depth causes eye fatigue and has reduced 3D effect.
Fig. 6 A-6C illustrates according to one embodiment of present invention and produces the stable and stereoscopic treatment of picture with being moved to the left to the right respectively by left-eye image and eye image with Fig. 5 A-5C.In other words, Fig. 6 A-6C illustrates by adjusting parallax the image processing method of stablizing 3D rendering is provided to the user.Fig. 6 A illustrates the left-eye image among the accompanying drawing 5A that moves on to the right, and Fig. 6 B illustrates the eye image among the accompanying drawing 5B that moves on to the left side.Accompanying drawing 6C illustrates according to the eye image among the synthetic Fig. 6 A of the left-eye image with among Fig. 6 B of embodiments of the invention on the 3D three-dimensional display.
With reference to figure 6C, the whole parallax of synthetic image is less than the parallax of the superimposed images shown in the accompanying drawing 5C.Therefore, the image of some target, for example, the image 61 of target has negative parallax.Therefore in other words, image 61 has the negative degree of depth and looks like and is absorbed in screen.Image 64 is the convergence points that have parallax free and have the degree of depth that equals screen intensity.Image 62 and 63 parallax are positive.Therefore, image 62 and 63 looks like from screen outstanding.
When eye image and left-eye image synthesized shown in Fig. 6 C, the image of composograph internal object had the degree of depth more various than the superimposed images shown in Fig. 5 C.Therefore, can provide stable and three-dimensional more image to the user.Describe now a kind of according to an embodiment of the invention method of definite horizontal movement value, its right eye by estimating input when right eye and left-eye image are flatly mobile with predetermined space and the parallax between the left-eye image are determined the parallax histogram of the relation between horizontal movement value and the definite parallax frequency of occurrences of representing the parallax estimated in the image and estimation.
Fig. 7 illustrates according to an embodiment of the invention the disparity estimation (DE) based on piece (block-based).
With reference to the accompanying drawings 7, left-eye image is divided into the NxN piece of equal sizes.Use absolute difference and (SAD) or absolute difference average (MAD) estimate to be similar to most the piece of the eye image of corresponding piece in the left-eye image.In this case, the distance definition between benchmark (reference) piece and the estimation piece is disparity vector (DV).Usually, DV distributes to each pixel in the benchmark image.Yet, in order to reduce the amount of required calculating, approximate identical among the DV that supposes all pixels in the piece and the block-based DE.For carrying out DE to each pixel, the DV that obtains each pixel is known as DE based on pixel.
Block-based DE or be used for estimating disparity based on the DE of pixel.
Fig. 8 illustrates the histogram of parallax between the left-eye image and eye image according to an embodiment of the invention.With reference to the accompanying drawings 8, the trunnion axis of histogram is represented the parallax size, and the vertical axis of histogram is represented frequency,, has the number of unit (unit) image of identical big or small parallax that is.Cell picture represent among the block-based DE cell block and based on the unit pixel among the DE of pixel.
In the situation of the histogram shown in Fig. 8, the stable and stereoscopic image can offer user's eyes, and target seems outstanding from screen.In the histogram of Fig. 8, most of images have negative parallax, and other has a positive parallax.
Therefore, the distribution of the histogram of general stereo-picture need be adjusted to the histogram distribution of Fig. 8.
Fig. 9 A-9C illustrates three kinds of histograms of parallax between the left-eye image in the multi-view image that obtains by parallel many view camera and eye image according to an embodiment of the invention.There are various types of parallax histograms in type and some stereo-picture parameters (such as the distance between camera and the focal length) according to stereoscopic camera.The histogram of parallax can be categorized as three kinds of histograms shown in Fig. 9 A-9C between left-eye image by parallel many view cameras photograph and the eye image.The histogram of Fig. 9 A presents than higher frequency in the low zone of parallax.The zone of the histogram of Fig. 9 B in parallax is inclined to one side presents than higher frequency.The histogram of Fig. 9 C presents than higher frequency in the high zone of parallax.
Yet, as Fig. 9 A in the horizontal direction of parallel many views shown in the 9C or three-dimensional view all parallax value usually greater than zero, this be because camera aligning parallel to each other and therefore optical axis do not converge on any one point.This left eye and eye image with this histogram can cause serious eye fatigue.Stabilized frequency distribution map shown in Fig. 8 can reduce eye fatigue and be used to produce the 3D stereo-picture.
According to one embodiment of present invention, Fig. 9 A is adjusted to the hypothetical parallax histogram shown in Fig. 8 to the 3D stereo-picture histogram shown in the 9C.For this reason, accompanying drawing 9A need move up at losing side to the histogram shown in the 9C.
As mentioned above, in order to produce the input picture of the parallax histogram that has shown in Fig. 8, analyze the parallax histogram of input picture for each incoming frame, and calculate parallax threshold value Dth, described threshold values makes the estimated rate (a%) of all parallaxes in the parallax histogram of input picture for just.In this case, a is the steady state value of experimentally determining.Persons skilled in the art can be known parallax threshold value Dth and can use several different methods to calculate.
The threshold value of calculating in the histogram of Fig. 9 A is Dth1, is Dth2 in the histogram of Fig. 9 B, and is Dth3 in the histogram of Fig. 9 C.When accompanying drawing 9A to the histogram of 9C respectively by the threshold value Dth1 that calculates to Dth3 when losing side moves up, (that is a%) will be as shown in Figure 8 positive parallax, at the estimated rate of Fig. 9 A total parallax in each histogram of 9C.
In the parallax adjustment is handled, move right and eye image is moved to the left based on the threshold value left-eye image of calculating.Then, the left-eye image and the eye image that move horizontally are synthesized.The position that optimal depth consciousness is provided that is used for the convergence point of target is placed stereoscopic screen is handled in this parallax adjustment.
Figure 10 A-10C illustrates according to an embodiment of the invention based on the left-eye image of horizontal movement value and moving horizontally of eye image, and described horizontal movement value uses the parallax histogram to determine.
As mentioned above, according to present embodiment, by the parallax of parallax threshold value Dth adjustment 3D rendering, so that positive parallax occupies the a% of parallax total between left eye and the eye image.10A-10C with reference to the accompanying drawings, left-eye image moves horizontally Dth/2 to the right, and eye image moves horizontally Dth/2 left.In this case, the parallax of 3D stereo-picture is adjusted.
Figure 10 C illustrate flatly move and overlapping Figure 10 A in left-eye image and the eye image among Figure 10 B.With reference to figure 10C, superimposed images internal object 101 has negative parallax, and target 102 and 103 has positive parallax, and target 104 has parallax free.Therefore, when using the parallax of the parallax adjustment 3D rendering between left eye and the eye image, the target in the superimposed images has the different degree of depth.Therefore, can provide stable and three-dimensional more image to the user.
Figure 11 is the block diagram according to embodiments of the invention treatment of picture equipment 100.With reference to Figure 11, image processing equipment 100 comprises image division unit 11, histogram generation unit 13, horizontal movement value determining unit 15, horizontal movement unit 17 and user interface section 19.
Image division unit 11 receives 3D rendering and 3D rendering is divided into eye image and left-eye image.Then, image division unit 11 sends eye image and left-eye image to histogram generation unit 13 and horizontal movement unit 17.
The parallax that histogram generation unit 13 is estimated between eye image and the left-eye image is measured the frequency that estimating disparity takes place, and produces the histogram for parallax and frequency.In this case, can use block-based as mentioned above DE or based on DE or other method of pixel.
Horizontal movement value determining unit 15 receives the histogram that produces from histogram generation unit 13, and the horizontal movement value of definite left eye and eye image.At length, horizontal movement value determining unit 15 is analyzed the parallax histogram of each incoming frame, calculates parallax threshold value Dth, (it makes positive parallax occupy the estimated rate of total parallax, that is, a%), and determine that corresponding to half value of the parallax threshold value Dth that calculates be the horizontal movement value.
Horizontal movement value determining unit 15 can be passed through the estimated rate that user interface section 19 receives from the user, and definite parallax threshold value Dth.Perhaps, horizontal movement value determining unit 15 can directly receive parallax threshold value Dth from the user by user interface section 19.In this case, the user can import different ratios or threshold value experimentally, determines suitable horizontal movement value, and adjusts the parallax of 3D rendering so that optimum 3D effect to be provided based on the threshold value of determining.
Horizontal movement unit 17 moves right left-eye image and eye image is moved to the left the horizontal movement value that receives from horizontal movement value determining unit 15, and the left eye and the eye image of output adjusted parallax.Though not shown, parallax is adjusted equipment 100 can comprise that the 3D stereoscopic display unit is to show the left eye and the eye image of the adjusted parallax of 17 outputs from the horizontal movement unit.
Obviously left-eye image and eye image needn't move equal amount.For example, eye image can be moved, and left-eye image is not moved.
Figure 12 illustrates image processing method flow chart according to an embodiment of the invention.With reference to Figure 12,3D rendering is received and is divided into left-eye image and eye image (S120).Estimate the parallax between left-eye image and the eye image, measure the frequency of the parallax generation of estimating, and produce histogram (S122) for parallax and frequency.Determine the horizontal movement value (S124) of eye image and left-eye image based on histogram.At length, calculate the parallax threshold value, it makes the estimated rate of all parallaxes in the parallax histogram for just, and the parallax threshold value that use is calculated is determined the horizontal movement value.The horizontal movement value can be defined as equaling half of threshold value.In addition, estimated rate or parallax threshold value can be imported as user input signal by the user.
Left-eye image and eye image are horizontally moved based on the horizontal movement value of determining, and adjust the parallax (S126) between left-eye image and the eye image.The left eye and the eye image of output and demonstration adjusted parallax.
Image processing method according to an aspect of the present invention and device can be used for controlling based on the parallax between left eye and the eye image convergence of the image that obtains by many view camera, and therefore described image can be used for 3D stereo display, thereby strengthens 3D effect and reduce eye fatigue.
Image processing method according to an aspect of the present invention and device can be used for by receiving and showing the convergence (convergence) that the parallax of the receiving terminal adjustment image of 3D rendering comes the control chart picture.
The present invention can also be embodied as the embodied on computer readable code on the computer-readable medium storing.In addition, being used to finish function program of the present invention, code and code segment can be easily be explained by the technical program person in the field under the present invention.Computer-readable medium storing is any data storage device that can store thereafter by the data of computer system reads.The example of computer-readable medium storing comprises read-only memory (ROM), random-access memory (ram), CD-ROM, tape, floppy disk, optical data storage equipment and carrier wave (transmitting such as the data by the Internet).Computer-readable medium storing can also be that distributed network coupled computers system stores and carries out with distributed way so that embodied on computer readable is encoded.
Though illustrate and described the present invention especially in conjunction with exemplary embodiment wherein, yet one of ordinary skill in the art is understandable that not breaking away from the defined the spirit and scope of the present invention of claims can make different variation on form and the details.

Claims (12)

1.一种图像处理设备,包括:1. An image processing device, comprising: 图像划分单元,其接收三维图像并且将三维图像划分为左眼图像和右眼图像;an image division unit that receives a three-dimensional image and divides the three-dimensional image into a left-eye image and a right-eye image; 频率分布图产生单元,其估计左眼图像和右眼图像的相应单元图像之间的视差,测量估计的视差发生的频率,并且产生视差和频率的频率分布图;a frequency histogram generating unit that estimates disparity between corresponding unit images of the left-eye image and the right-eye image, measures frequency at which the estimated disparity occurs, and generates a frequency histogram of the disparity and frequency; 水平运动值确定单元,其使用产生的频率分布图确定左眼图像和右眼图像的水平运动值;以及a horizontal motion value determination unit that determines horizontal motion values of the left-eye image and the right-eye image using the generated frequency histogram; and 水平运动单元,其基于确定的水平运动值水平地移动左眼图像和右眼图像中的至少一个,a horizontal motion unit that horizontally moves at least one of the left-eye image and the right-eye image based on the determined horizontal motion value, 其中,所述水平运动值确定单元计算阈值,该阈值使得频率分布图中的视差的预定比率为正视差,并且使用计算的阈值来确定水平运动值,wherein the horizontal motion value determination unit calculates a threshold that makes a predetermined ratio of parallax in the frequency histogram a positive parallax, and determines the horizontal motion value using the calculated threshold, 其中,所述预定比率为被确定为使得移动左眼图像和右眼图像中的至少一个后三维图像具有最优深度知觉的恒定值。Wherein, the predetermined ratio is a constant value determined such that the three-dimensional image after moving at least one of the left-eye image and the right-eye image has optimal depth perception. 2.权利要求1的设备,其中,所述水平运动值确定单元确定对应于阈值一半的值为水平运动值。2. The apparatus of claim 1, wherein the horizontal motion value determination unit determines a value corresponding to half of the threshold value as the horizontal motion value. 3.权利要求2的设备,其中,所述水平运动单元基于所述阈值水平地移动左眼图像和右眼图像,并且,因此调整视差。3. The apparatus of claim 2, wherein the horizontal motion unit horizontally moves the left-eye image and the right-eye image based on the threshold, and adjusts the disparity accordingly. 4.权利要求1的设备,进一步包括用户接口单元,其通过用户输入信号从用户接收预定比率或阈值并且将预定比率或阈值传送给水平运动值确定单元,其中,所述水平运动值确定单元使用接收的预定比率或阈值来确定水平运动值。4. The apparatus of claim 1, further comprising a user interface unit that receives a predetermined ratio or threshold from a user via a user input signal and transmits the predetermined ratio or threshold to a horizontal motion value determination unit, wherein the horizontal motion value determination unit uses A predetermined ratio or threshold is received to determine a horizontal motion value. 5.权利要求1的设备,进一步包括三维立体显示单元,其显示从水平运动单元输出的已调整视差的左眼图像和右眼图像。5. The apparatus of claim 1, further comprising a three-dimensional stereoscopic display unit displaying the parallax-adjusted left-eye image and right-eye image output from the horizontal motion unit. 6.一种图像处理方法,包括:6. An image processing method, comprising: 接收三维图像并且将三维图像划分为左眼图像和右眼图像;receiving a three-dimensional image and dividing the three-dimensional image into a left-eye image and a right-eye image; 估计左眼图像和右眼图像的相应单元图像之间的视差,测量估计视差发生的频率,并且产生视差和频率的频率分布图;estimating a disparity between corresponding unit images of the left-eye image and the right-eye image, measuring a frequency at which the estimated disparity occurs, and generating a frequency map of the disparity and frequency; 使用产生的频率分布图来确定左眼图像和右眼图像的水平运动值;以及using the resulting histogram to determine horizontal motion values for the left-eye image and the right-eye image; and 基于确定的水平运动值水平地移动左眼图像和右眼图像中的至少一个以调整视差,horizontally moving at least one of the left-eye image and the right-eye image to adjust disparity based on the determined horizontal motion value, 其中,所述水平运动值的确定包括:Wherein, the determination of the horizontal motion value includes: 计算阈值,该阈值使得频率分布图中视差的预定比率为正视差;以及calculating a threshold such that a predetermined ratio of disparities in the histogram is positive disparity; and 使用所计算的阈值确定水平运动值,Determine the horizontal motion value using the calculated threshold, 其中,所述预定比率为被确定为使得移动左眼图像和右眼图像中的至少一个后三维图像具有最优深度知觉的恒定值。Wherein, the predetermined ratio is a constant value determined such that the three-dimensional image after moving at least one of the left-eye image and the right-eye image has optimal depth perception. 7.权利要求6的方法,其中,所述水平运动值的确定包括确定对应于阈值一半的值为水平运动值。7. The method of claim 6, wherein determining the horizontal motion value includes determining a value corresponding to half the threshold value as a horizontal motion value. 8.权利要求7的方法,其中,所述左眼图像和右眼图像的水平运动包括基于阈值水平地移动左眼图像和右眼图像,并且,因此调整视差。8. The method of claim 7, wherein the horizontal movement of the left-eye image and the right-eye image comprises horizontally moving the left-eye image and the right-eye image based on a threshold, and adjusting parallax accordingly. 9.权利要求6的方法,进一步的包括:9. The method of claim 6, further comprising: 通过用户输入信号接收来自用户的预定比率或阈值;以及receiving a predetermined ratio or threshold from a user via a user input signal; and 使用接收的预定比率或阈值来确定水平运动值。A horizontal motion value is determined using the received predetermined ratio or threshold. 10.权利要求6的方法,进一步包括显示已调整视差的左眼图像和右眼图像。10. The method of claim 6, further comprising displaying the parallax-adjusted left-eye and right-eye images. 11.一种图像处理设备,包括:11. An image processing device comprising: 图像划分单元,其接收图像并且将图像划分为第一图像和第二图像;an image division unit that receives an image and divides the image into a first image and a second image; 频率分布图产生单元,其估计第一图像和第二图像的相应单元图像之间的视差,测量估计的视差发生的频率,并且产生视差和频率的频率分布图;a frequency histogram generating unit that estimates disparity between corresponding unit images of the first image and the second image, measures the frequency at which the estimated disparity occurs, and generates a frequency histogram of the disparity and frequency; 运动值确定单元,其使用产生的频率分布图确定第一图像和第二图像的运动值;以及a motion value determination unit that determines motion values of the first image and the second image using the generated frequency histogram; and 运动单元,其基于确定的运动值来移动第一图像和第二图像中的至少一个,并且,因此调整图像的视差,a motion unit that moves at least one of the first image and the second image based on the determined motion value and, accordingly, adjusts the disparity of the images, 其中,所述运动值的确定包括:Wherein, the determination of the motion value includes: 计算阈值,该阈值使得频率分布图中视差的预定比率为正视差;以及calculating a threshold such that a predetermined ratio of disparities in the histogram is positive disparity; and 使用所计算的阈值确定所述运动值,determining said motion value using the calculated threshold, 其中,所述预定比率为被确定为使得移动第一图像和第二图像中的至少一个后获得的三维图像具有最优深度知觉的恒定值。Wherein, the predetermined ratio is a constant value determined such that a three-dimensional image obtained after moving at least one of the first image and the second image has optimal depth perception. 12.一种图像处理方法,包括:12. An image processing method, comprising: 接收图像,并且将图像划分为第一图像和第二图像;receiving an image, and dividing the image into a first image and a second image; 估计第一图像和第二图像的相应单元图像之间的视差,测量估计的视差发生的频率,并且产生视差和频率的频率分布图;estimating a disparity between corresponding unit images of the first image and the second image, measuring a frequency at which the estimated disparity occurs, and generating a frequency histogram of the disparity and frequency; 使用产生的频率分布图确定第一图像和第二图像的运动值;以及determining motion values for the first image and the second image using the generated histogram; and 基于确定的运动值移动第一图像和第二图像中的至少一个,并且,因此调整图像的视差,moving at least one of the first image and the second image based on the determined motion value, and, accordingly, adjusting the disparity of the images, 其中,所述运动值的确定包括:Wherein, the determination of the motion value includes: 计算阈值,该阈值使得频率分布图中视差的预定比率为正视差;以及calculating a threshold such that a predetermined ratio of disparities in the histogram is positive disparity; and 使用所计算的阈值确定所述运动值,determining said motion value using the calculated threshold, 其中,所述预定比率为被确定为使得移动第一图像和第二图像中的至少一个后获得的三维图像具有最优深度知觉的恒定值。Wherein, the predetermined ratio is a constant value determined such that a three-dimensional image obtained after moving at least one of the first image and the second image has optimal depth perception.
CN2006101639357A 2005-10-12 2006-10-12 Apparatus and method for processing 3d picture Expired - Fee Related CN1956555B (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR96196/05 2005-10-12
KR1020050096196A KR101185870B1 (en) 2005-10-12 2005-10-12 Apparatus and method for processing 3 dimensional picture

Publications (2)

Publication Number Publication Date
CN1956555A CN1956555A (en) 2007-05-02
CN1956555B true CN1956555B (en) 2011-12-07

Family

ID=37911112

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2006101639357A Expired - Fee Related CN1956555B (en) 2005-10-12 2006-10-12 Apparatus and method for processing 3d picture

Country Status (4)

Country Link
US (1) US8116557B2 (en)
KR (1) KR101185870B1 (en)
CN (1) CN1956555B (en)
NL (1) NL1032656C2 (en)

Families Citing this family (54)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101311896B1 (en) * 2006-11-14 2013-10-14 삼성전자주식회사 Displacement adjustment method of stereoscopic image and stereoscopic image device applying the same
KR101345303B1 (en) * 2007-03-29 2013-12-27 삼성전자주식회사 Dynamic depth control method or apparatus in stereo-view or multiview sequence images
KR101176065B1 (en) * 2008-12-22 2012-08-24 한국전자통신연구원 Method for transmitting data on stereoscopic image, method for playback of stereoscopic image, and method for creating file of stereoscopic image
JP5274359B2 (en) 2009-04-27 2013-08-28 三菱電機株式会社 3D video and audio recording method, 3D video and audio playback method, 3D video and audio recording device, 3D video and audio playback device, 3D video and audio recording medium
US9524700B2 (en) * 2009-05-14 2016-12-20 Pure Depth Limited Method and system for displaying images of various formats on a single display
JP2011030182A (en) * 2009-06-29 2011-02-10 Sony Corp Three-dimensional image data transmission device, three-dimensional image data transmission method, three-dimensional image data reception device, and three-dimensional image data reception method
KR20110005205A (en) * 2009-07-09 2011-01-17 삼성전자주식회사 Signal processing method and apparatus using screen size of display device
JP5249149B2 (en) * 2009-07-17 2013-07-31 富士フイルム株式会社 Stereoscopic image recording apparatus and method, stereoscopic image output apparatus and method, and stereoscopic image recording and output system
KR20110018261A (en) * 2009-08-17 2011-02-23 삼성전자주식회사 Text subtitle data processing method and playback device
JP5444955B2 (en) * 2009-08-31 2014-03-19 ソニー株式会社 Stereoscopic image display system, parallax conversion device, parallax conversion method, and program
CN102033414B (en) * 2009-09-27 2012-06-20 深圳市掌网立体时代视讯技术有限公司 Stereoscopic digital imaging converging device and method
US8284235B2 (en) * 2009-09-28 2012-10-09 Sharp Laboratories Of America, Inc. Reduction of viewer discomfort for stereoscopic images
US8798160B2 (en) 2009-11-06 2014-08-05 Samsung Electronics Co., Ltd. Method and apparatus for adjusting parallax in three-dimensional video
JP5756119B2 (en) * 2009-11-18 2015-07-29 トムソン ライセンシングThomson Licensing Method and system for 3D content distribution with flexible parallax selection
WO2011081642A1 (en) * 2009-12-14 2011-07-07 Thomson Licensing Image pair processing
JP2011166285A (en) * 2010-02-05 2011-08-25 Sony Corp Image display device, image display viewing system and image display method
JP2011176541A (en) * 2010-02-24 2011-09-08 Sony Corp Three-dimensional video processing apparatus and method, and program thereof
WO2011114739A1 (en) * 2010-03-17 2011-09-22 パナソニック株式会社 Replay device
US8213708B2 (en) * 2010-03-22 2012-07-03 Eastman Kodak Company Adjusting perspective for objects in stereoscopic images
US9001227B2 (en) 2010-04-05 2015-04-07 Qualcomm Incorporated Combining data from multiple image sensors
US8896668B2 (en) 2010-04-05 2014-11-25 Qualcomm Incorporated Combining data from multiple image sensors
JP5669599B2 (en) * 2010-05-14 2015-02-12 キヤノン株式会社 Image processing apparatus and control method thereof
US8970672B2 (en) * 2010-05-28 2015-03-03 Qualcomm Incorporated Three-dimensional image processing
KR101685343B1 (en) * 2010-06-01 2016-12-12 엘지전자 주식회사 Image Display Device and Operating Method for the Same
EP2586209A1 (en) 2010-06-28 2013-05-01 Thomson Licensing Method and apparatus for customizing 3-dimensional effects of stereo content
JP5623156B2 (en) * 2010-06-29 2014-11-12 キヤノン株式会社 Image processing apparatus and control method thereof
JP2012044383A (en) * 2010-08-18 2012-03-01 Sony Corp Image processing device, method and program
US8767053B2 (en) * 2010-08-26 2014-07-01 Stmicroelectronics, Inc. Method and apparatus for viewing stereoscopic video material simultaneously with multiple participants
KR20120030639A (en) * 2010-09-20 2012-03-29 삼성전자주식회사 Display apparatus and image processing method of the same
US8947511B2 (en) 2010-10-01 2015-02-03 At&T Intellectual Property I, L.P. Apparatus and method for presenting three-dimensional media content
FR2967324B1 (en) * 2010-11-05 2016-11-04 Transvideo METHOD AND DEVICE FOR CONTROLLING THE PHASING BETWEEN STEREOSCOPIC CAMERAS
KR20120051308A (en) * 2010-11-12 2012-05-22 삼성전자주식회사 Method for improving 3 dimensional effect and reducing visual fatigue and apparatus of enabling the method
CN102006493A (en) * 2010-11-26 2011-04-06 北京新岸线网络技术有限公司 Parallax adjustment method and device for 3D video image
US9049423B2 (en) * 2010-12-01 2015-06-02 Qualcomm Incorporated Zero disparity plane for feedback-based three-dimensional video
WO2012075603A1 (en) * 2010-12-08 2012-06-14 Technicolor (China) Technology Co., Ltd. Method and system for 3d display with adaptive disparity
KR101296902B1 (en) * 2010-12-08 2013-08-14 엘지디스플레이 주식회사 Image processing unit and stereoscopic image display device using the same, and image processing method
US9088835B2 (en) * 2010-12-17 2015-07-21 Thomson Licensing Method for adjusting depth or view of three-dimensional streaming video
US9591281B2 (en) * 2010-12-22 2017-03-07 Thomson Licensing Apparatus and method for determining a disparity estimate
JP5002066B1 (en) * 2011-04-06 2012-08-15 シャープ株式会社 Image processing apparatus, image processing method, computer program, and recording medium
KR101777875B1 (en) * 2011-04-28 2017-09-13 엘지디스플레이 주식회사 Stereoscopic image display and method of adjusting stereoscopic image thereof
US20130063576A1 (en) * 2011-04-28 2013-03-14 Panasonic Corporation Stereoscopic intensity adjustment device, stereoscopic intensity adjustment method, program, integrated circuit and recording medium
JP2012257022A (en) * 2011-06-08 2012-12-27 Sony Corp Image processing apparatus, method, and program
TWI486052B (en) * 2011-07-05 2015-05-21 Realtek Semiconductor Corp Three-dimensional image processing device and three-dimensional image processing method
WO2013027343A1 (en) * 2011-08-23 2013-02-28 パナソニック株式会社 Three-dimensional image capture device, lens control device, and program
KR101872864B1 (en) * 2011-12-19 2018-06-29 엘지전자 주식회사 Electronic device and controlling method for electronic device
TWI514849B (en) * 2012-01-11 2015-12-21 Himax Tech Ltd Calibration device used in stereoscopic display system and calibration method of the same
KR102143944B1 (en) * 2013-09-27 2020-08-12 엘지디스플레이 주식회사 Method of adjusting three-dimensional effect and stereoscopic image display using the same
TWI577493B (en) * 2014-12-26 2017-04-11 財團法人工業技術研究院 Calibration method and automatic apparatus using the same
CA3021538C (en) * 2015-06-11 2023-09-26 University Of Pittsburgh-Of The Commonwealth System Of Higher Education Systems and methods for finding regions of interest in hematoxylin and eosin (h&e) stained tissue images and quantifying intratumor cellular spatial heterogeneity in multiplexed/hyperplexed fluorescence tissue images
US9529454B1 (en) 2015-06-19 2016-12-27 Microsoft Technology Licensing, Llc Three-dimensional user input
KR101917762B1 (en) * 2017-05-22 2019-01-29 주식회사 그루크리에이티브랩 Method for reduction of eye-fatigue while using a virtual reality apparatus by controlling convergence
WO2019041035A1 (en) 2017-08-30 2019-03-07 Innovations Mindtrick Inc. Viewer-adjusted stereoscopic image display
JP7517339B2 (en) * 2019-08-08 2024-07-17 ソニーグループ株式会社 Surgery image display system, image processing device, and image processing method
CN119596551B (en) * 2024-12-25 2025-12-02 歌尔光学科技有限公司 Smart glasses and their parallax correction methods, as well as augmented reality devices

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5867591A (en) * 1995-04-21 1999-02-02 Matsushita Electric Industrial Co., Ltd. Method of matching stereo images and method of measuring disparity between these image
CN1638491A (en) * 2003-11-13 2005-07-13 三星电子株式会社 Temporal smoothing apparatus and method for synthesizing intermediate image
CN1643939A (en) * 2002-03-27 2005-07-20 三洋电机株式会社 Method and apparatus for processing three-dimensional images

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5020878A (en) * 1989-03-20 1991-06-04 Tektronix, Inc. Method and apparatus for generating a binocular viewing model automatically adapted to a selected image
JP3242529B2 (en) 1994-06-07 2001-12-25 松下通信工業株式会社 Stereo image matching method and stereo image parallax measurement method
JP2002125245A (en) 2000-10-17 2002-04-26 Sony Corp Stereoscopic image editing method, stereoscopic image editing apparatus, and stereoscopic image display system
KR100392381B1 (en) 2000-11-13 2003-07-22 한국전자통신연구원 Method and apparatus controlling the disparity of objects in images and a parallel-axis stereo camera system using the apparatus
JP2003018619A (en) * 2001-07-03 2003-01-17 Olympus Optical Co Ltd Three-dimensional image evaluation apparatus and display using the same
JP2003158751A (en) 2001-11-20 2003-05-30 Fuji Heavy Ind Ltd Stereo camera adjustment device
JP2005073049A (en) 2003-08-26 2005-03-17 Sharp Corp 3D image reproduction apparatus and 3D image reproduction method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5867591A (en) * 1995-04-21 1999-02-02 Matsushita Electric Industrial Co., Ltd. Method of matching stereo images and method of measuring disparity between these image
CN1643939A (en) * 2002-03-27 2005-07-20 三洋电机株式会社 Method and apparatus for processing three-dimensional images
CN1638491A (en) * 2003-11-13 2005-07-13 三星电子株式会社 Temporal smoothing apparatus and method for synthesizing intermediate image

Also Published As

Publication number Publication date
NL1032656A1 (en) 2007-04-13
US20070081716A1 (en) 2007-04-12
NL1032656C2 (en) 2011-06-06
US8116557B2 (en) 2012-02-14
KR20070040645A (en) 2007-04-17
CN1956555A (en) 2007-05-02
KR101185870B1 (en) 2012-09-25

Similar Documents

Publication Publication Date Title
CN1956555B (en) Apparatus and method for processing 3d picture
EP2532166B1 (en) Method, apparatus and computer program for selecting a stereoscopic imaging viewpoint pair
RU2431938C2 (en) Efficient multiple types encoding
CN1956554B (en) Apparatus and method for processing 3d picture
US8045792B2 (en) Method and apparatus for controlling dynamic depth of stereo-view or multi-view sequence images
JP5654138B2 (en) Hybrid reality for 3D human machine interface
JP6027034B2 (en) 3D image error improving method and apparatus
Schmeing et al. Depth image based rendering: A faithful approach for the disocclusion problem
US20140218490A1 (en) Receiver-Side Adjustment of Stereoscopic Images
Kim et al. Depth adjustment for stereoscopic image using visual fatigue prediction and depth-based view synthesis
US10554954B2 (en) Stereoscopic focus point adjustment
Knorr et al. From 2D-to stereo-to multi-view video
US9693042B2 (en) Foreground and background detection in a video
US9591290B2 (en) Stereoscopic video generation
JP5871113B2 (en) Stereo image generation apparatus, stereo image generation method, and stereo image generation program
US9674500B2 (en) Stereoscopic depth adjustment
US9661309B2 (en) Stereoscopic video zooming
Kang et al. Efficient disparity map generation for moving multi-camera system using TOF depth sensor
Kaller et al. Hyper stereo base cameras setting for widescreen long distance target shooting application

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20111207

Termination date: 20211012

CF01 Termination of patent right due to non-payment of annual fee